1 Genome Sourcing

Initially 212 Corynebacterium assemblies were collected from the following sources:

1. 16 genomes labelled as KPL that correspond to isolates originally sequenced as part of the HMP. These were deposited as Corynebacterium sp. prior to this study by the Broad Institute under PRJNA169439.

2. 94 genomes from strains labelled as KPL and isolated in MA as part of this study and with a putative 16S rRNA gene assignment to a Corynebacterium spp. Deposited at NHI databases under PRJNA842433.

3. 71 genomes from strains labelled as BWA and isolated in Botswana as part of this study. Deposited at NHI databases under XXXXX.

4. 31 reference assemblies for the nasal Corynebacterium species of interest. They were downloaded using using PanACoTa with the PanACoTA prepare -s subcommand searching by the name of the species, as written by the NCBI. Default parameters were used in order to discard:

   • low quality assemblies: maximum value of L90 allowed = 100 and maximum number of contigs allowed to keep a genome = 999.
   • redundant pairs or mislabeled assemblies: min MASH (Ondov et al. 2016) distance to the reference = 1e-4 and max MASH (Ondov et al. 2016) distance to the reference = 0.06.

    PanACoTA prepare -s "Corynebacterium <SPECIES_NAME>" 

2 Genome Selection

PanACoTA prepare --norefseq -d "<GENOMES_FOLDER>" -o "<OUTPUT_FOLDER>" --max 1 -M